One of the most basic and probably one of the simplest dimensionality reduction techniques called Principal component analysis. Dimensionality reduction basically, what it does is, if your data points lie in d dimensional space and if you want to transform your data points Xi’s , Xi belongs to Rd . We have to reduce d dimensions to d’ dimensions such that d’<d, which is called as dimensionality reduction. For MNIST dataset, we have 784 dimensions; if we convert this data in to two dimensions then visualization is very easy. We will apply linear algebraic transformations to convert data from higher dimensions to lower dimension and which is much easier to interpret and understand. We collected MNIST dataset from Kaggle and Kaggle is the phenomenal repository of great datasets

PCA, MNIST dataset, Dimensionality reduction, Geometrical interpretation, Mathematical Interpretation